The alamios/Mistral-Small-3.1-DRAFT-0.5B is a 0.5 billion parameter draft model designed for speculative decoding. It is specifically intended to accelerate inference when paired with larger Mistral-Small-3.1-24B-Instruct or Mistral-Small-24B-Instruct models. This model was trained on Mistral's outputs, covering a variety of tasks and datasets in English, French, German, Spanish, Italian, and Portuguese, making it suitable for multilingual applications.
No reviews yet. Be the first to review!